scholarly journals Concept Discovery for The Interpretation of Landscape Scenicness

2020 ◽  
Vol 2 (4) ◽  
pp. 397-413
Author(s):  
Pim Arendsen ◽  
Diego Marcos ◽  
Devis Tuia
Keyword(s):  
Vector Spaces ◽  
Large Set ◽  
Feature Representations ◽  
The United Kingdom ◽  
New Concepts ◽  
Manifold Alignment ◽  
Semantic Concepts ◽  
Visual Concepts ◽  
Alignment Technique ◽  
Concept Activation

In this paper, we study how to extract visual concepts to understand landscape scenicness. Using visual feature representations from a Convolutional Neural Network (CNN), we learn a number of Concept Activation Vectors (CAV) aligned with semantic concepts from ancillary datasets. These concepts represent objects, attributes or scene categories that describe outdoor images. We then use these CAVs to study their impact on the (crowdsourced) perception of beauty of landscapes in the United Kingdom. Finally, we deploy a technique to explore new concepts beyond those initially available in the ancillary dataset: Using a semi-supervised manifold alignment technique, we align the CNN image representation to a large set of word embeddings, therefore giving access to entire dictionaries of concepts. This allows us to obtain a list of new concept candidates to improve our understanding of the elements that contribute the most to the perception of scenicness. We do this without the need for any additional data by leveraging the commonalities in the visual and word vector spaces. Our results suggest that new and potentially useful concepts can be discovered by leveraging neighbourhood structures in the word vector spaces.

Creative Modernity: The New Cultural State

2004 ◽  
Vol 112 (1) ◽  
pp. 9-27 ◽  
Author(s):  
Steve Redhead
Keyword(s):  
Culture Media ◽  
Social Engineering ◽  
Creative Industries ◽  
Theory And Practice ◽  
Modern State ◽  
International Debate ◽  
The United Kingdom ◽  
New Concepts ◽  
The Social ◽  
Cultural Sphere

This essay introduces two new concepts into the international debate about the theory and practice of creative industries. These concepts are ‘creative modernity’ and the ‘new cultural state’. The new cultural state has a double meaning. It refers to the new cultural condition we find ourselves in, what we call here creative modernity, and the form in which the modern state has governed, or intervened in, culture through law and other means of governance or regulation. In this process, the modern state — as it did in the United Kingdom for a while — sometimes becomes a part of the ‘cultural’ sphere through the project of creative modernity. As we see here in a rethinking of the case of the Department of Culture, Media and Sport in the British New Labour government, an experiment which is often cited approvingly in the creative industries debates around the world, creative modernity involves the social engineering of a ‘new individualism’ where citizens are remade as creative entrepreneurs. In this essay, it is argued that to move the arguments forward, the debate about creative industries should be re-situated within the wider framework of cosmopolitan sociology's analyses of modernity, the state and culture.

MMIR

2011 ◽  
pp. 371-387 ◽  
Author(s):  
Min Chen ◽  
Shu-Ching Chen
Keyword(s):  
Multiple Instance Learning ◽  
Image Features ◽  
Large Set ◽  
Hierarchical Learning ◽  
Retrieval Accuracy ◽  
The Neural Network ◽  
Semantic Concepts ◽  
Major Bottleneck ◽  
Speed Up ◽  
High Level

This chapter introduces an advanced content-based image retrieval (CBIR) system, MMIR, where Markov model mediator (MMM) and multiple instance learning (MIL) techniques are integrated seamlessly and act coherently as a hierarchical learning engine to boost both the retrieval accuracy and efficiency. It is well-understood that the major bottleneck of CBIR systems is the large semantic gap between the low-level image features and the high-level semantic concepts. In addition, the perception subjectivity problem also challenges a CBIR system. To address these issues and challenges, the proposed MMIR system utilizes the MMM mechanism to direct the focus on the image level analysis together with the MIL technique (with the neural network technique as its core) to real-time capture and learn the object-level semantic concepts with some help of the user feedbacks. In addition, from a long-term learning perspective, the user feedback logs are explored by MMM to speed up the learning process and to increase the retrieval accuracy for a query. The comparative studies on a large set of real-world images demonstrate the promising performance of our proposed MMIR system.

General anaesthesia for caesarean delivery

2016 ◽  
Author(s):  
David M. Levy ◽  
Ieva Saule
Keyword(s):  
United Kingdom ◽  
Laryngeal Mask Airway ◽  
General Anaesthesia ◽  
Caesarean Delivery ◽  
Laryngeal Mask ◽  
World Health ◽  
Induction Agent ◽  
Large Set ◽  
The United Kingdom ◽  
Airway Trauma

General anaesthesia (GA) is most often indicated for category 1 (immediate threat to life of mother or baby) caesarean delivery (CD) or when neuraxial anaesthesia has failed or is contraindicated. Secure intravenous access is essential. Jugular venous cannulation (with ultrasound guidance) is required if peripheral access is inadequate. A World Health Organization surgical safety checklist must be used. The shoulders and upper back should be ramped. Left lateral table tilt or other means of uterine displacement are essential to minimize aortocaval compression, and a head-up position is recommended to improve the efficiency of preoxygenation and reduce the likelihood of gastric contents reaching the oropharynx. Cricoid pressure is controversial. In the United Kingdom, thiopental remains the induction agent of choice, although there is scant evidence upon which to avoid propofol. In pre-eclampsia, it is essential to obtund the pressor response to laryngoscopy with remifentanil or alfentanil. Rocuronium is an acceptable alternative to succinylcholine for neuromuscular blockade. Sugammadex offers the possibility of swifter reversal of rocuronium than spontaneous recovery from succinylcholine. Management of difficult tracheal intubation is focused on ‘oxygenation without aspiration’ and prevention of airway trauma. The Classic™ laryngeal mask airway is the most commonly used rescue airway in the United Kingdom. There is a large set of data from fasted women of low body mass index who have undergone elective CD safely with a Proseal™ or Supreme™ laryngeal mask airway. Sevoflurane is the most popular volatile agent for maintenance of GA. The role of electroencephalography-based depth of anaesthesia monitors at CD remains to be established. Intraoperative end-tidal carbon dioxide tension should be maintained below 4.0 kPa.

Social Pedagogy: Differences and Links to Existing Child Care Practice

2011 ◽  
Vol 36 (4) ◽  
pp. 199-206 ◽  
Author(s):  
Robyn Kemp
Keyword(s):  
Child Care ◽  
Good Practice ◽  
Care Practice ◽  
Original Form ◽  
Pedagogical Approach ◽  
The Third ◽  
The United Kingdom ◽  
Social Pedagogy ◽  
New Concepts ◽  
The Uk

Some commentators have pointed to the United Kingdom (UK) having a tendency towards reducing ‘new’ concepts or practices so much that they bear little resemblance to the original form. This is why I wish to highlight in this article some of the sometimes subtle, yet profound, differences in a social pedagogical approach to child care in order to better understand the potential of social pedagogy for developing practice. There are five main sections to this article: first, I describe social pedagogy and a conceptualisation of a social pedagogical approach; second, the UK context is examined so as to set the scene for the third, fourth and fifth sections, which examine reflection, relationships and the concept of lifespace through a social pedagogic lens, drawing links to existing good practice in the UK. Although this commentary does not discuss Australian practice, I believe there are some important aspects in the way in which the UK has tried to familiarise itself with social pedagogy that can help Australian readers to better understand some of the subtleties and nuances of the paradigm and inspire their own reflections.

scholarly journals Social Impact Assessment: A Systematic Review of Literature

2021 ◽  
Author(s):  
William Alomoto ◽  
Angels Niñerola ◽  
Laia Pié
Keyword(s):  
Social Impact ◽  
Social Impact Assessment ◽  
The Sustainable Development ◽  
The United Kingdom ◽  
Current State ◽  
New Concepts ◽  
The Usa ◽  
Social Environmental ◽  
Large Clusters ◽  
Determining Factor

AbstractMeasuring, analyzing, and evaluating social, environmental, and economic impact is crucial to aligning the sustainable development strategies of international organizations, governments, and businesses. In this sense, society has been a determining factor exerting pressure for urgent solutions. The main objective of this paper is to provide an exhaustive analysis of the literature about the tools for measuring social impact and their evolution over the last 50 years. The search was conducted in the main academic databases (Scopus and Web of Science), where 924 articles were found from 1969 to 2020 related to the topic. The results of the quantitative analysis show that 71% of the publications were in the last ten years and the most productive countries were the USA and the United Kingdom. The relational analysis identifies 4 large clusters that fragment the literature into different subfields. The most used keywords are linked to the term "Social" in measurement methods, new concepts, and participants. This article contributes to the literature by giving the researcher an insight into the current state of art, trends, categories within the field, and future lines of research.

MMIR

2009 ◽  
pp. 1189-1204
Author(s):  
Min Chen ◽  
Shu-Ching Chen
Keyword(s):  
Multiple Instance Learning ◽  
Image Features ◽  
Large Set ◽  
Hierarchical Learning ◽  
Retrieval Accuracy ◽  
The Neural Network ◽  
Semantic Concepts ◽  
Major Bottleneck ◽  
Speed Up ◽  
High Level

This chapter introduces an advanced content-based image retrieval (CBIR) system, MMIR, where Markov model mediator (MMM) and multiple instance learning (MIL) techniques are integrated seamlessly and act coherently as a hierarchical learning engine to boost both the retrieval accuracy and efficiency. It is well-understood that the major bottleneck of CBIR systems is the large semantic gap between the low-level image features and the high-level semantic concepts. In addition, the perception subjectivity problem also challenges a CBIR system. To address these issues and challenges, the proposed MMIR system utilizes the MMM mechanism to direct the focus on the image level analysis together with the MIL technique (with the neural network technique as its core) to real-time capture and learn the objectlevel semantic concepts with some help of the user feedbacks. In addition, from a long-term learning perspective, the user feedback logs are explored by MMM to speed up the learning process and to increase the retrieval accuracy for a query. The comparative studies on a large set of real-world images demonstrate the promising performance of our proposed MMIR system.

MMIR

2011 ◽  
pp. 103-120
Author(s):  
Min Chen ◽  
Shu-Ching Chen
Keyword(s):  
Multiple Instance Learning ◽  
Image Features ◽  
Large Set ◽  
Hierarchical Learning ◽  
Retrieval Accuracy ◽  
The Neural Network ◽  
Semantic Concepts ◽  
Major Bottleneck ◽  
Speed Up

This chapter introduces an advanced content-based image retrieval (CBIR) system, MMIR, where Markov model mediator (MMM) and multiple instance learning (MIL) techniques are integrated seamlessly and act coherently as a hierarchical learning engine to boost both the retrieval accuracy and efficiency. It is well-understood that the major bottleneck of CBIR systems is the large semantic gap between the low-level image features and the highlevel semantic concepts. In addition, the perception subjectivity problem also challenges a CBIR system. To address these issues and challenges, the proposed MMIR system utilizes the MMM mechanism to direct the focus on the image level analysis together with the MIL technique (with the neural network technique as its core) to real-time capture and learn the object-level semantic concepts with some help of the user feedbacks. In addition, from a long-term learning perspective, the user feedback logs are explored by MMM to speed up the learning process and to increase the retrieval accuracy for a query. The comparative studies on a large set of real-world images demonstrate the promising performance of our proposed MMIR system.

scholarly journals A Non-Linear Manifold Alignment Approach to Robot Learning from Demonstrations

2018 ◽  
Vol 30 (2) ◽  
pp. 265-281 ◽  
Author(s):  
Ndivhuwo Makondo ◽  
Michihisa Hiratsuka ◽  
Benjamin Rosman ◽  
Osamu Hasegawa ◽  
◽  
...  
Keyword(s):  
Linear Manifold ◽  
Robot Learning ◽  
Procrustes Analysis ◽  
Human Model ◽  
Dynamic Movement ◽  
Learning From Demonstrations ◽  
Manifold Alignment ◽  
Alignment Technique ◽  
Dynamic Movement Primitives ◽  
Skeletal Model

The number and variety of robots active in real-world environments are growing, as well as the skills they are expected to acquire, and to this end we present an approach for non-robotics-expert users to be able to easily teach a skill to a robot with potentially different, but unknown, kinematics from humans. This paper proposes a method that enables robots with unknown kinematics to learn skills from demonstrations. Our proposed method requires a motion trajectory obtained from human demonstrations via a vision-based system, which is then projected onto a corresponding human skeletal model. The kinematics mapping between the robot and the human model is learned by employing Local Procrustes Analysis, a manifold alignment technique which enables the transfer of the demonstrated trajectory from the human model to the robot. Finally, the transferred trajectory is encoded onto a parameterized motion skill, using Dynamic Movement Primitives, allowing it to be generalized to different situations. Experiments in simulation on the PR2 and Meka robots show that our method is able to correctly imitate various skills demonstrated by a human, and an analysis of the transfer of the acquired skills between the two robots is provided.

scholarly journals A Large Set of Designs On Vector Spaces

1994 ◽  
Vol 47 (3) ◽  
pp. 247-272 ◽  
Author(s):  
D.K. Raychaudhuri ◽  
E.J. Schram
Keyword(s):  
Vector Spaces ◽  
Large Set

scholarly journals Interpreting and Comparing Convolutional Neural Networks: A Quantitative Approach

2021 ◽  
Author(s):  
Mohammad Mohaiminul Islam ◽  
Zahid Hassan Tushar
Keyword(s):  
Neural Network ◽  
Region Of Interest ◽  
Input Image ◽  
Black Box ◽  
Discriminative Ability ◽  
Highly Active ◽  
Semantic Concepts ◽  
Gradient Based ◽  
Visual Concepts ◽  
Hidden Neurons

A convolutional neural network (CNN) is sometimes understood as a black box in the sense that while it can approximate any function, studying its structure will not give us any insights into the nature of the function being approximated. In other terms, the discriminative ability does not reveal much about the latent representation of a network. This research aims to establish a framework for interpreting the CNNs by profiling them in terms of interpretable visual concepts and verifying them by means of Integrated Gradient. We also ask the question, "Do different input classes have a relationship or are they unrelated?" For instance, could there be an overlapping set of highly active neurons to identify different classes? Could there be a set of neurons that are useful for one input class whereas misleading for a different one? Intuition answers these questions positively, implying the existence of a structured set of neurons inclined to a particular class. Knowing this structure has significant values; it provides a principled way for identifying redundancies across the classes. Here the interpretability profiling has been done by evaluating the correspondence between individual hidden neurons and a set of human-understandable visual semantic concepts. We also propose an integrated gradient-based class-specific relevance mapping approach that takes the spatial position of the region of interest in the input image. Our relevance score verifies the interpretability scores in terms of neurons tuned to a particular concept/class. Further, we perform network ablation and measure the performance of the network based on our approach.

Sign in / Sign up

Export Citation Format

Share Document